​
Overview

Adaptive is a drop-in replacement for OpenAI that automatically routes your requests to the best AI model for each task. Simply change your base URL and start saving up to 80% on AI costs.
Zero code changes required - Works with any OpenAI-compatible client by just changing the base URL.

​
Step 1: Get Your API Key

1

Sign Up

Create a free account at llmadaptive.uk
2

Create Project

Set up a new project in your dashboard
3

Generate API Key

Generate your API key from the API Keys section
Keep your API key secure. Never expose it in client-side code or public repositories.

​
Step 2: Update Your Code

The only change needed is updating your base URL: 
import OpenAI from 'openai';

const openai = new OpenAI({
  apiKey: 'your-adaptive-api-key',
  baseURL: 'https://llmadaptive.uk/api/v1'
});

// Use exactly like OpenAI
const completion = await openai.chat.completions.create({
  model: '', // Leave empty for intelligent routing
  messages: [
    { role: 'user', content: 'Explain quantum computing simply' }
  ],
});

console.log(completion.choices[0].message.content);

​
Step 3: Test Your Integration

Run your code to see Adaptive in action:

​
Key Differences from OpenAI

Base URL

OpenAI: https://api.openai.com/v1
Adaptive: https://llmadaptive.uk/api/v1

Model Selection

OpenAI: Specify exact model
Adaptive: Use "" for intelligent routing

Providers

OpenAI: OpenAI only
Adaptive: 6+ providers (OpenAI, Anthropic, Google, etc.)

Cost

OpenAI: Fixed pricing
Adaptive: Up to 80% savings

​
Advanced Configuration

​
Control Provider Selection

Limit which providers Adaptive can choose from: 
const completion = await openai.chat.completions.create({
  model: '',
  messages: [{ role: 'user', content: 'Hello!' }],
  provider_constraints: ['openai', 'anthropic'], // Only use these
});

​
Cost vs Performance Balance

Control the cost/performance trade-off: 
const completion = await openai.chat.completions.create({
  model: '',
  messages: [{ role: 'user', content: 'Hello!' }],
  cost_bias: 0.2, // 0 = cheapest, 1 = best performance
});

​
Streaming Support

Streaming works exactly like OpenAI: 
const stream = await openai.chat.completions.create({
  model: '',
  messages: [{ role: 'user', content: 'Tell me a story' }],
  stream: true,
});

for await (const chunk of stream) {
  process.stdout.write(chunk.choices[0]?.delta?.content || '');
}

​
Environment Setup

For production deployments: 
# .env file
ADAPTIVE_API_KEY=your-adaptive-api-key
ADAPTIVE_BASE_URL=https://llmadaptive.uk/api/v1

const openai = new OpenAI({
  apiKey: process.env.ADAPTIVE_API_KEY,
  baseURL: process.env.ADAPTIVE_BASE_URL
});

​
Error Handling

Handle errors the same way as OpenAI: 
try {
  const completion = await openai.chat.completions.create({
    model: '',
    messages: [{ role: 'user', content: 'Hello!' }],
  });
} catch (error) {
  if (error.status === 401) {
    console.error('Invalid API key');
  } else if (error.status === 429) {
    console.error('Rate limit exceeded');
  } else {
    console.error('Request failed:', error.message);
  }
}

​
Verification Checklist

​
Next Steps

Integration Guides

Framework-specific guides for OpenAI SDK, Vercel AI, LangChain

Features

Learn about intelligent routing and cost optimization

API Reference

Complete API documentation and all parameters

Examples

Working code examples for common use cases

​
Need Help?

Support

Email us at support@adaptive.com

Migration Guide

See our migration guide for detailed steps